Title | ||
---|---|---|
SAM-Net: Semantic probabilistic and attention mechanisms of dynamic objects for self-supervised depth and camera pose estimation in visual odometry applications |
Abstract | ||
---|---|---|
•We propose a self-supervised learning-based VO framework: SAM-Net.•We present SPFM to detect dynamic objects and to generate refined depth map.•Apply attention mechanism to enhance perception ability in spatial and channel view.•PoseNet with the atrous separable convolution is designed to expand receptive field.•Photometric consistency loss is more robust for the large rotations. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1016/j.patrec.2021.11.028 | Pattern Recognition Letters |
Keywords | DocType | Volume |
Visual odometry,Self-supervised deep learning,Object detection,Semantic probabilistic map,Attention mechanism | Journal | 153 |
ISSN | Citations | PageRank |
0167-8655 | 0 | 0.34 |
References | Authors | |
0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Binchao Yang | 1 | 0 | 0.34 |
Xinying Xu | 2 | 0 | 0.34 |
Jinchang Ren | 3 | 1144 | 88.54 |
Lan Cheng | 4 | 0 | 0.34 |
Lei Guo | 5 | 0 | 0.34 |
Zhe Zhang | 6 | 0 | 0.34 |